SIOC English
化学学报
高级检索

化学学报 ›› 2008, Vol. 66 ›› Issue (7): 793-798. 上一篇    下一篇

研究论文

一种新颖的基于FTIR-CWT及ANN分类法的同科属中药材菟丝子与金灯藤的识别研究

程存归 田玉梅 张长江   

  1. 浙江师范大学
  • 投稿日期:2007-10-22 修回日期:2008-01-16 发布日期:2008-04-14
  • 通讯作者: 程存归

A Novel Recognition Method Between Semen Cuscutae and Japanese Dodder Seed Based on FTIR-Continuous-Wavelet Feature Extraction and Artificial Neural Network Classification Method

  

  • Received:2007-10-22 Revised:2008-01-16 Published:2008-04-14

PDF

545

采用水平衰减全反射傅里叶变换红外光谱法（HATR-FTIR）测定了同属种子植物中药材菟丝子及金灯藤的FTIR，运用基于连续小波多分辨率分析法对吸收较为相似的菟丝子及金灯藤的FTIR进行特征提取。选择第7、10、13分解层数的特征向量，进行人工神经网络（ANN）训练，再用训练出来的网络对不同产地的植物种子菟丝子和金灯藤所得FTIR小波提取的特征向量进行分类。通过对32个不同样本的验证，说明能够采用基于FTIR-连续小波特征提取及人工神经网络分类法对同科属中药材菟丝子与金灯藤进行识别。

Fourier Transform Infrared (FTIR) and Horizontal Attenuated Total Reflectance (HATR) techniques were used to obtain the FTIR of semen cuscutae (the seed from Cuscuta chinensis lam) and Japanese dodder seed (the seed from Cuscuta japonica Choisy). The similar features between the semen cuscutae and Japanese dodder seed are extracted by continuous wavelet transform. The scale 7, 10 and 13 are used to extract the feature vectors, which are used to train the artificial neural network（ANN）. The trained neural network is used to classify semen cuscutaes and Japanese dodder seeds, which are collected from different places all over the country.According to 32 testing samples, we could effectively identify the sibling plants, semen cuscutae and Japanese dodder seed by FTIR with continuous wavelet feature extraction and artificial neural network classification.